Research On Preform Optimal Design Based On Bi-directional Evolutionary Structural Optimization

Forgings with complex shapes are difficult to be forged directly.It generally needs one or more preform steps to obtain an eligible final forging.Preform design is the most important and difficult task in forging process design.A rational preform design is the key to ensuring a good forming quality of forging.With the development of computer technology and finite element analysis(FEA)technology,domestic and foreign scholars have proposed various preform optimal design methods based on forming process simulation.Based on analysis of the advantages and disadvantages of the various preform optimal design methods,this thesis proposes a new method of preform optimal design that applies the bi-directional evolutionary structural optimization(BESO)to the preform design of forging process.The BESO is a novel optimal method originally applied to the continuum structural optimal design.It obtains the optimal structural shape that meets the requirements of mechanical properties by gradually adding or deleting materials according to certain addition/deletion criterion.There are many differences between preform optimal design of material plastic forming and structural optimal design of continuum.Therefore,the BESO cannot be directly applied to the preform optimal design.After analyzing the differences between the preform optimal design and structural optimal design,this thesis improves the methods of BESO,and proposes the concrete realization scheme and implementation flow of the preform optimal design based on BESO.According to the characteristics of the preform optimal design,the improvement schemes of objective functions,the background grid system,the addition/deletion criterion,and the element deletion restrictions are proposed.In this thesis,a series of key techniques of preform optimal design based on BESO are studied in detail,including objective functions,element addition/deletion criterion,element information tracking and transferring,element addition and deletion and smooth fitting of boundary.Taking the completely filling of cavity and uniform deformation of forging as the optimization goals,and the mathematical models of the total objective function and sub-objective functions are established.A new element addition/deletion criterion is proposed that is associated with the objective functions.The new addition/deletion criterion will automatically adjust its value based on the current value of objective functions,which has certain self-adaptability.The implementation method of element information tracking and transferring is proposed,which can track and transfer information between finite element mesh and background grid.The specific steps of the element addition and deletion are given,and a new method of boundary smoothing fitting is proposed.The programs for preform optimal design has been developed by C++.The key techniques used in preform optimal design including grid discretization of initial preform billets,generation of background grid,DEFORM data read of FEA,calculation of objective functions,calculation of value of element addition/deletion criterion,element information tracking and transferring,element addition and deletion and smooth fitting of boundary et al.In order to verify the reliability of the proposed preform optimal design method,the H-shaped forging and axisymmetric disk forging are taken as typical examples to carry out the preform optimal design.Complete filling of die cavity and uniform deformation of the forging are the goals of preform optimal design.Building the optimal analysis model for the two examples and carring out the FEA by using DEFORM-2D.After several preform optimization iterations,the ideal results of preform optimal design are obtained.